JP5135628B2 - Oscillator and oscillator using the same - Google Patents

Description

  The present invention relates to an oscillator using MEMS technology and an oscillator using the same.

There is an increasing demand for miniaturization and high accuracy of various electronic components such as portable information terminal devices typified by cellular phones, etc., and such portable information terminal devices are small and stable high-frequency signal sources. Is indispensable.
A typical electronic component for satisfying this requirement is a crystal resonator. A quartz resonator is known to have a very high resonance sharpness (ie, Q value), which is an index of the quality of an oscillator, because of good crystal stability. This is the reason why crystal resonators are widely used as stable high-frequency signal sources for wireless portable devices and personal computers.
However, it has become clear that this quartz crystal resonator cannot sufficiently satisfy the recent demand for further downsizing.

Therefore, in recent years, a microscale mechanical resonator having a high Q value using a vibrator formed by MEMS (Micro-Electro-Mechanical-System) technology using silicon instead of a quartz vibrator has been developed. . Recent advances in micromachining have enabled the reduction in size and cost of highly stable oscillators used in communications and clock (timekeeper) applications.
In particular, as a surface micromechanical resonator made of silicon, an IC-compatible oscillator that has a Q value exceeding 10,000 and can oscillate between MF and VHF frequencies has been demonstrated to work with polycrystalline silicon as a structural material. (For example, refer nonpatent literature 1).

A micromechanical oscillator in which the resonator and the peripheral circuits that maintain this oscillation are integrated on a single chip using planar processing combining surface micromachining and integrated circuits has also been demonstrated to operate at a high Q value. (For example, see Non-Patent Document 2).
CT-c.Nguyen, “Frequency-Selective MEMS for Miniaturized Low-Power Communication Devices (invited),” IEEE Trans. MICROWAVE THEORY TECH., Vol.47, No.8, pp.1486-1503, August 1999 CT-c.Nguyen and RTHowe, “An Integrated CMOS Micromechanical Resonator High-Q Oscillator,” IEEE SOLID-STATE CIRCUITS, Vol.34, No.4, pp.440-445, April 1999

However, in the above-described mechanical oscillator, for example, as illustrated in FIG. 5, in the resonator 32, the protrusions 40 a and 40 b of the fixed electrodes 33 a and 33 b (drive electrode and detection electrode) are opposed to the vibrating piece 36. Has a gap, and the mechanical resonance frequency is set by the mechanical dimension, Young's modulus and density of the resonator element 36. Capacitance elements are formed by these gaps.
Further, in addition to the opposing portions that form the capacitance, the parasitic capacitances in the fixed electrodes 33a and 33b and the vibrator islands 34a and 34b that fix both ends of the oscillation piece 36 so that the oscillation piece 36 can vibrate are described above. It is added to the capacitance element.
Therefore, in the mechanical oscillator, for example, when the driving electrode is the fixed electrode 33a and the detection electrode is the fixed electrode 33b, the capacitance change due to the change in the distance between the opposing detection electrode 33b and the vibrator is detected in the oscillation state. However, when the drive signal is applied to the drive electrode 33a, the crosstalk due to the increase in the parasitic capacitance from the drive electrode 33a to the detection electrode 33b is detected. Is included in the detection signal.

As a result, the conventional mechanical oscillator has a drawback that when it is driven to oscillate, the influence of parasitic capacitance is large and the Q value is lowered, and stable oscillation cannot be performed.
In addition, as described above, when another circuit is formed on the same substrate, the operation of the other circuit causes the influence of the parasitic capacitance existing between the fixed electrodes 33a and 33b and the other circuit. Interference with the amplifier amplification circuit occurs through the detection electrode 33b, so that stable oscillation cannot be performed and abnormal oscillation occurs.
The present invention has been made in view of such circumstances, and the parasitic capacitance between the fixed electrode and the vibrator other than the capacitive coupling that sets the resonance frequency is determined with respect to the resonator composed of the fixed electrode and the vibrator in the mechanical oscillator. An object of the present invention is to provide an oscillator that suppresses and stably oscillates and an oscillator using the same.

  The resonator according to the present invention includes a vibrator having an end fixed and a long shape, and a fixed electrode having a facing portion facing the vibrator in a direction perpendicular to the longitudinal direction of the vibrator. And a shield provided in a region where the fixed electrode and the vibrator face each other and having a window portion at a portion where the facing portion faces the vibrator.

  In the resonator according to the aspect of the invention, the shield may be provided on the entire outer peripheral portion of the fixed electrode except for the window portion.

  In the resonator according to the aspect of the invention, the shield may be provided on the entire outer peripheral portion of the vibrator excluding the window portion.

  An oscillator according to the present invention includes the resonator according to any one of claims 1 to 3 and an oscillation circuit that oscillates at a resonance frequency of the resonator and outputs an oscillation signal. .

  The oscillator according to the present invention is characterized in that a preset potential is applied to the shield of the oscillator.

As described above, according to the present invention, since the shield in which only the portion of the vibrator and the fixed electrode facing each other is opened as a window is provided between the vibrator and the fixed electrode, the resonance frequency Since only the portion that determines the capacitance value for setting the capacitance is capacitively coupled and shields the electric field other than the opposing portion that contributes to the setting of the resonance frequency, formation of parasitic capacitance that causes crosstalk to the detection electrode by the drive signal And a stable oscillator having a high Q value can be provided.
The effect is obtained.

Further, according to the present invention, interference from other circuits formed on the same substrate with respect to the amplifier circuit of the oscillator can be suppressed, and an abnormal oscillation can be prevented and a stable oscillator can be provided.
The effect is obtained.

Hereinafter, an oscillator according to an embodiment of the present invention will be described with reference to the drawings. FIG. 1 is a perspective view showing a configuration example of an oscillator using the resonator according to the embodiment. 2 shows the structure of the oscillator constituting the oscillator of FIG. 1, FIG. 2 (a) is a plan view, and FIG. 2 (b) is a cross section taken along the line AA 'of FIG. FIG. The same components as those in the conventional example of FIG.
In the present embodiment, an oscillator used for various electronic components such as a mobile phone and a portable information terminal device will be described. Moreover, in each figure shown below, in order to make each layer and each member the size which can be recognized on drawing, the scale is varied for each layer and each member.

In FIG. 1, an oscillator 30 includes a silicon support layer 11, a BOX (Buried Oxide) layer 12 of silicon dioxide (SiO ₂ , hereinafter referred to as an oxide film), and a so-called SOI (stacked SOI) layer. It is manufactured by a semiconductor process technology (for example, MEMS technology) using a silicon-on-insulator substrate 45.
However, the oscillator 30 may be manufactured not only with the SOI substrate 45 but also with a semiconductor substrate such as silicon. Among these layers, the silicon active layer 44 includes a frame 31 that rises along the outer periphery of the silicon support layer 11, and a vibrator 32 and a pair of fixed electrodes 33 a and 33 b disposed inside the frame 31. ing. The fixed electrodes 33a and 33b and the vibrator 32 are formed on the BOX layer 12 by a conductor such as silicon (for example, the silicon active layer 44). Formed).
In the present embodiment, the vibrator 32 is constituted by a double-supported beam so that the vibration piece 36 can freely vibrate. However, the vibrator 32 may have a cantilever structure in which one is fixed.

As described above, the IC substrate 20 is placed on the BOX layer 12 laminated on the silicon support layer 11, and the vibrator 36 and the fixed electrodes 33 a and 33 b of the oscillator 30 and the bonding wires 52, 53 and 55 are mounted. Thus, the electrode pads of the IC circuit 21 are electrically connected.
In the present embodiment, the IC substrate 20 and the oscillator 30 are integrally packaged via the silicon support layer 11.

Accordingly, the silicon support layer 11 functions as a common substrate that also supports the IC substrate 20 at the same time. The IC substrate 20 is placed with a slight gap with respect to the frame 31.
As shown in FIG. 2A, the vibrator 32 has a substantially I shape in plan view, and is based on the vibrator islands 34a and 34b (base portion) and the vibrator islands 34a and 34b. And a vibrating piece 36 supported in a cantilever manner via end portions 35a and 35b.

The vibrator islands 34 a and 34 b have a rectangular shape in plan view, and are formed on the silicon support layer 11 via the BOX layer 12. The vibrator island 34 a is configured so that a preset voltage is applied to the vibrator 32 by being electrically connected to the electrode pad of the IC 21 through the bonding wire 53.
The vibrating piece 36 is supported by the vibrating portion islands 34a and 34b at the base end portions 35a and 35b, and vibrates in the X direction (one direction) orthogonal to the side surface 34d along the longitudinal direction of the vibrator island 34a. It extends to the child island 34b. The vibrating piece 36 extends with a gap between the lower surface 36a and the upper surface 11a of the silicon support layer 11 (see FIG. 2B), and is orthogonal to the extending direction (X direction). Is configured to vibrate in the Y direction (the width direction (other direction) of the resonator element 36).

  On both sides of the vibrator 32, a pair of fixed electrodes are sandwiched between the vibrators 32 in a state perpendicular to the longitudinal direction of the vibrator with a gap d between the vibrators 32. Parts 33a and 33b are arranged. The pair of fixed electrode portions 33 a and 33 b are formed along the width direction of the resonator element 36. Each of the fixed electrode portions 33 a and 33 b includes convex portions 40 a and 40 b (opposing portions) facing the side surfaces 41 a and 41 b in the longitudinal direction of the vibrating piece 36, and the protruding portions 40 a and 40 b and the vibrating piece 36 Opposing portions have a gap d. Here, the side surface 41a faces the convex portion 40a, and the side surface 41b faces the convex portion 40b. The vibrator 32 and the convex portions 40a, 40a, and 40b are formed according to the width and height (opposite area) of the side surfaces 41a and 41b on both sides of the vibrating piece 36 and the side surfaces 48a and 48b of the convex portions 40a and 40b. The capacitance value of the capacitive element by 40b is set. Here, the capacitance element indicates a region where the side surfaces 41a and 41b on both sides of the vibrating piece 36 are opposed to the side surfaces 48a and 48b of the convex portions 40a and 40b.

  The shield 50 has a wall-like structure formed of a conductor (for example, a metal such as aluminum or polycrystalline silicon doped with impurities), and includes vibrator islands 34a and 34b, electrode portions 33a, A gap portion facing 33b, that is, a region where the transducer islands 34a and 34b and the electrode portions 33a and 33b face each other is formed. The thickness of the shield 50 in the direction of the fixed electrode and the transducer island is smaller than the gap between the fixed electrode and the transducer island. Further, in a region other than the region where the vibration piece 36 vibrates (that is, the region where the vibration piece 36 (including the base end portions 35a and 35b) and the projections 40a and 40b face each other), that is, the region where the shield 50 is formed. An insulating film such as an oxide film is formed between the shield 50 and the fixed electrode and the vibrator island, and the shield 50 and the fixed electrode and the vibrator island are electrically insulated.

  Here, portions where the side surfaces 41 a and 41 b of the vibrating piece 36 and the side surfaces 48 a and 48 b of the convex portions 40 a and 40 b face each other are opened as a window of the shield 50. The shield 50 is electrically connected to the electrode pad of the IC 21 by the electrode 51 through the bonding wire 54, and a preset voltage is applied from a low impedance power source or the like. Thereby, the electric field generated between the fixed electrode on which the shield 50 is formed and the vibrator island is shielded, and the parasitic capacitance in this region can be suppressed. As a result, for example, when the fixed electrode 33a is used as a driving electrode and the fixed electrode 33b is used as a detection electrode, a driving voltage for driving the vibrating piece 36 by electrostatic attraction or electrostatic repulsion is applied to the driving electrode. Since the electric field due to the drive voltage is shielded, crosstalk due to the drive voltage with respect to the detection electrode can be reduced. The potential applied to the shield 50 will be described later.

Further, as shown in FIG. 2A, the shield 50 has a pattern 1 configuration formed as a partial shield 50a in a gap between regions where the transducer islands 34a and 34b and the fixed electrode portions 33a and 33b face each other.
Further, as a combination of the partial shield 50a and the partial shield 50b, there is a configuration of a pattern 2 that surrounds the entire electrode portions 33a and 33b except for portions where the side surfaces 48a and 48b of the convex portions 40a and 40b are opened as windows. .
Further, as a combination of the partial shield 50a, the partial shield 50b, and the partial shield 50c, the pattern 3 surrounding the electrode portions 33a and 33b and the transducer islands 34a and 34b, except for the portions where the side surfaces 48a and 48b are opened as windows. There is a configuration.

Any of the configurations of the pattern 1 to the pattern 3 described above may be used, but the effect of suppressing the crosstalk differs depending on the patterns 1, 2, and 3.
For example, the case where the fixed electrode 33a is a drive electrode for the vibrating piece 36 and the fixed electrode 33b is a detection electrode for detecting the vibration of the vibrating piece 36 will be described.
In the oscillator 10, an oscillation circuit (described later) of the IC 21 detects the vibration of the vibrating piece 36 detected by the detection electrode 33b, and the oscillation voltage of the IC 21 (amplifier circuit in the IC 21) amplifies the detected voltage. A drive voltage is applied to 33a, and the vibrating piece 36 biased by the bias voltage is driven by electrostatic attraction and electrostatic repulsion.
At this time, the voltage applied to the drive electrode 33a is crosstalked with respect to the transducer islands 34a and 34b capacitively coupled to the drive electrode 33a (having parasitic capacitance) and the detection electrode 33b. Will occur, resulting in an unstable oscillation state.

However, in the present embodiment, since the shield of the pattern 1 is adopted, the facing portion between the driving electrode 33a and the transducer islands 34a and 34b, and the facing portion between the transducer islands 34a and 34b and the detection electrode 33b. Since the electric field is shielded and the largest capacitive coupling is suppressed, the crosstalk can be reduced as compared with the conventional case.
When the shield of the pattern 2 is employed, the electric field that reaches the transducer islands 34a and 34b and the detection electrode 33b from the drive electrode 33a to the transducer islands 34a and 34b and the detection electrode 33b through the shield 50a. Can be shielded, and an electric field from an external circuit (IC21 or the like) can be shielded, and crosstalk can be further suppressed with respect to the pattern 1.
When the shield of the pattern 3 is adopted, the electric field from the external circuit to the vibrator islands 34 a and 34 b can be shielded in addition to the pattern 2, and the crosstalk is further suppressed with respect to the patterns 1 and 2. be able to.

Further, as already described, the fixed electrode portions 33a and 33b are formed on the silicon support layer 11 with the BOX layer 12 interposed therebetween, and the side surfaces 48a and 48b thereof correspond to the side surfaces 41a and 41b of the resonator element 36. The gap 32 is formed so as to surround the side of the vibrator 32 while having a gap d in parallel.
The oscillator 10 is usually provided with a sealing substrate so as to cover the oscillator 30 in order to maintain the inside of the frame 31 of the oscillator 30 in a vacuum state. In this embodiment, the description of the sealing substrate is omitted.

Next, the operation of the oscillator 10 of this embodiment will be described with reference to FIGS.
In the following description, the fixed electrode 33a is used as a drive electrode for the vibrating piece 36, and the fixed electrode 33b is used as a detection electrode for detecting vibration of the vibrating piece 36. However, either of the fixed electrodes 33a and 33b is used as a detection electrode. A drive electrode may be used.
The oscillation circuit 22 includes inverters B1, B2, and B3, load resistors R1, R2, and R3, and capacitors C1 and C2.

The inverters B1, B2, and B3 are connected in series by inserting a capacitor C1 between the inverters B1 and B2 and a capacitor C2 between the inverters B2 and B3 as a DC component cutting capacitor. The load resistors R1, R2, and R3 are connected in parallel to the inverters B1, B2, and B3, respectively.
The input terminal of the inverter B3 becomes the input of the oscillation circuit 22, the output terminal of the inverter B1 becomes the output of the oscillation circuit 22, and positive feedback is applied at the oscillation frequency.
The output of the oscillation circuit 22 is connected to the drive electrode 33a via the wire bonding 52, and the input of the oscillation circuit 22 is connected to the detection electrode 33b via the wire bonding 55.

First, a voltage is applied to the IC 21 via an electric circuit from an external power source (not shown), which shows the oscillation circuit 22 formed in the IC 21 in FIG. 3, and a control signal is input.
Accordingly, the IC 21 applies the bias voltage Vb to the vibrator 32 via the bonding wire 53 in response to the control signal.
As a result, the oscillation circuit 22 amplifies a minute voltage fluctuation generated in the detection electrode 33b due to the unstable state of the vibrating piece 36 between the fixed electrode portions 33a and 33b, and drives the amplified voltage. By applying a positive feedback to the application electrode 33a, the frequency transmitted through the mechanical resonance portion becomes an oscillation state in several milliseconds.

When continuous oscillation is started, a stable drive voltage is applied to the drive electrode 33 a, and electrostatic attractive force or repulsive force is alternately generated between the vibrating piece 36.
As a result, each of the pair of fixed electrode portions 33a and 33b disposed in the Y direction (perpendicular to the longitudinal direction of the vibration piece 36), that is, on both sides of the vibrator 36 via the gap d is obtained as the vibration piece 36. It will vibrate so that it may approach and separate from the convex parts 40a and 40b. Here, the vibration piece 36 vibrates due to the electrostatic attraction and repulsion generated between the vibration piece 36 and the convex portion 40a by the driving voltage.
When the vibrating piece 36 vibrates, the gap between the vibrating piece 36 and the electrode portion 33b changes, and the capacitance of the capacitive element between the vibrating piece 36 and the detection electrode 33b changes. Then, the oscillation circuit 22 detects the change in capacitance of the capacitance element as a voltage change by the detection electrode 33b, inputs the detected voltage as a detection signal, and amplifies the voltage of the detection signal. The phase is shifted by 180 ° and output as a drive voltage to the drive electrode 33a.

At this time, a voltage is applied to the shield 50. As shown in FIG. 4, in this case, as shown in FIG. 4, in the case of FIG. 4A, the impedance of the ground (VS) or drive power supply (Vdd) for driving the IC 21 is obtained. In the case of FIG. 4B, it is directly connected to a power source having a low impedance of the voltage Vb applied to the vibrator 32, and in the case of FIG. 4C, the voltage of the detection electrode 33b is connected to the amplifier B4. Thus, there is a configuration in which power amplification (single voltage amplification) is performed and output to the shield 40.
In FIG. 4, the impedance element 200 inserted between the power source of the bias voltage and the vibrator island 34 a is, for example, a resistor or an inductor, and prevents inflow and outflow of a large current from the vibrator 32. Is provided to control the amount of current. In the case of FIG. 4C, the shield 50 is connected to a power source that directly outputs the bias Vb without passing through the impedance element 200.

As described above, in the present embodiment, when the configuration of the shield 50 is the pattern 1, the pattern 2, and the pattern 3, the electric field due to the driving voltage from the driving electrode 33a to the vibrator 32 and the detection electrode 33b is shielded, and the resonance The coupling capacitance (parasitic capacitance) other than the coupling capacitance that determines the frequency can be reduced, and the influence on the voltage detected by the lower detection electrode 33b compared to the drive voltage applied to the drive electrode 33a. Can be suppressed. For example, the drive voltage applied to the drive electrode 33a is several volts, and the voltage detected by the detection electrode 33b is several mV.
Further, when the configuration of the shield 50 is the pattern 2 and the pattern 3, the influence on the voltage detected by the detection electrode due to the driving of the circuit including the oscillation circuit 22 of the IC 21 can be suppressed.

1 is a perspective view illustrating a configuration example of an oscillator 10 according to an embodiment of the present invention. FIG. 2 is a diagram illustrating a configuration example of an oscillator 30 in FIG. 1. 1 is a block diagram illustrating a configuration example of an oscillator 10 using an oscillator 30 of the present embodiment. 3 is a conceptual diagram illustrating a configuration example in which a voltage is applied to a shield 50 in an oscillator 30. FIG. FIG. 6 is a plan view showing a configuration of a conventional oscillator 30.

Explanation of symbols

11 ... Silicon support layer 12 ... BOX layer 20 ... IC substrate 21 ... IC
DESCRIPTION OF SYMBOLS 22 ... Oscillator circuit 30 ... Oscillator 31 ... Frame 32 ... Vibrator 33a, 33b ... Fixing electrode 34a, 34b ... Vibrator island 35a, 35b ... Base end part 36 ... Vibrating piece 40a, 40b ... Protrusion part 41a, 41b, 48a, 48b ... side face 45 ... SOI substrate 50 ... shield 51 ... electrode 52, 53, 54, 55 ... bonding wire 200 ... impedance element B1, B2, B3 ... inverter B4 ... buffer C1, C2 ... capacitor

Claims (5)

An elongated vibrator in which the end of the vibrating piece is fixed;
A fixed electrode that is provided to face the vibrator in a direction perpendicular to the longitudinal direction of the vibrator and has a facing portion that faces the vibrating piece of the vibrator;
Wherein the fixed electrode is provided on the region opposed to the said oscillator, oscillator and the counter portion and the vibrating piece and a shield having a window portion at a portion facing.   The oscillator according to claim 1, wherein the shield is provided on the entire outer peripheral portion of the fixed electrode except for the window portion.   The oscillator according to claim 1, wherein the shield is provided on an entire outer peripheral portion of the vibrator excluding the window portion. The oscillator according to any one of claims 1 to 3,
And an oscillation circuit that oscillates at the resonance frequency of the resonator and outputs an oscillation signal.   The oscillator according to claim 4, wherein a preset potential is applied to the shield of the oscillator.

Images